Deep vein thrombosis (DVT) and pulmonary embolism (PE) most often occur during hospitalization for surgery or chronic medical conditions. Consequently information on DVT/PE is derived from large referral hospitals, and there are substantial gaps in our understanding of the incidence, risk factors, and outcomes of DVT/PE in the community. The proposed population-based study will identify the incidence cohort of Olmsted County Minnesota residents with DVT/PE from 1966 through 1990. Episodes of DVT or PE acquired during hospitalization or in the community will be identified for future studies. In a five part study we will: 1. identify the incidence cohort and test the hypotheses that over the 25 year study period the incidence of DVT/PE has decreased, the extent of testing for suspected DVT/PE has increased, and the number of tests obtained for patients with DVT/PE has increased. 2. determine the independent risk factors for DVT/PE and estimate the population attribute risk due to each independent risk factor for DVT/PE in a case-control study of the incidence cohort and community controls. 3. test the hypothesis that DVT/PE are independent risk factors for death after adjusting for age, sex and comorbid conditions. 4. describe the occurrence of the post-phlebitic syndrome, pulmonary hypertension, and recurrent DVT/PE over the 25 year study period and identify the independent risk factors for lower extremity venous ulceration. 5. evaluate the accuracy of national PE mortality rates by comparing PE mortality rates calculated from our review of all the inpatient and outpatient medical records of Olmsted County DVT/PE patients with death certificate PE mortality rates for Olmsted County from the National Center for Health Statistics. The study will provide the most complete and comprehensive population based estimates of the incidence, risk factors, survival and outcomes of DVT/PE available to date. This will be the first study to compare hospital acquired DVT/PE with DVT/PE among ambulatory residents in a community. With the identification of the population attribute risk due to each of the independent risk factors for DVT/PE and the prevalence cohort of Olmsted County residents with DVT/PE, further studies of mechanisms of thrombosis and interventions to prevent recurrence and improve long-term outcomes will be possible.